Skip to content

saketlab/seqout

Repository files navigation

Seqout

License GitHub Actions Workflow Status GitHub last commit

Seqout is a search engine for finding genomic datasets across NCBI and EBI portals.

Apart from text-based search, Seqout also offers dataset discovery via semantic similarity using vector embeddings. For example, check out: seqout.org/p/GSE153562#similar.

We also have a map of 800K+ datasets in a two-dimensional space, obtained via UMAP projection of the vector embeddings of the datasets. View it at seqout.org/map.

Seqout also has an MCP server to help work with genomic datasets using AI agents (such as Claude, Codex, Antigravity etc.). Visit seqout.org/map for more information.

Additionally, we also provide an enriched view of samples and experiments with standardized attributes. For example, check out: seqout.org/p/GSE44255#samples=enriched.